Raspje I was thinking in your line 9.dxc5 Ndxe5 10.Nxe5 Nxe5 11.Nb3?!  to play 11.Nf3 in stead with the idea 11..Nxd3 12.Qxd3 and the c-pawn wont fall as you have b4. I think white is better here.

Also 9.Bb5 directly looks good, which is a pity as I like the h5 idea.

Hello SmyslovFan,...and thanks for the reply.
Yes g5 is the other pawn move, and it's part of Black strategy.
Black want to play g5-g4 to force White to move his knight on f3 square defending the center. In main variations, Black try to push g5 at once but White has played dxc5 giving the d4 square for the knight,and Black has g4 pawn to protect after g5-g4. 
Of course, all the ideas for Black and White need "real" games or in-depth analysis. For example, I read from Eingorn's book that White want to play the f4 move but I have not found games analysis. 
The move f6 is key move for many lines in such pawn structures..and the question is: can Black afford such a move ? Without analysis, I guess the answer is NO if White has active development with light square bishop on d3 (the diag. e8-h5 is week) and no piece on e file will improve Re1 move. Hence, Black doesn't attack White pawns in center but pieces defending it.   
I can resist: I have already browsed Watson's book (Dangerous book The French) but not read from beginning to end..and I mention that the  positon of Diag 19 is about same topic except that h5 is only played after h3 for White

I finished reading Sveshnikov's books about the Advance and now I began reading  "Chess Explained - The French" Eingorn-Bogdanov Gambit Ed.

For general comment I would say book is an "overview" book, because it covers many main lines in few pages. Chess explained style is more dedicated to good comments about "reference games" than in-depth analysis.

In this thread I want just put in light, I found good analysis of Smirin-Cheparinov, Calvia Olympiad 2004, a Tarrasch variation game (what I call Geller/Korchnoi gambit pawn structure, but can be classify in Universal System too)

The comment that the h5 pawn move creates less weakness than g5, hence dxc5 is less stronger, makes me curious about using the h5 pawn move in another line.

I would like to know what you think about this (new) move:

1.e4 e6 2.d4 d5 3.Nd2 Nf6 4.e5 Nfd7 5.Ngf3 c5 6.c3 Nc6 (Korchnoi-Udovcic,Leningrad 1967 (Korchnoi,"My best games")) 7.Bd3 a5! (many purposes move: one common is to prepare vs a later Nb3 after the exchange cxd4-cxd4) 8.oo (8.a4 Pavasovic- A Jurkovic, 1999 ; 8.a3 Be7) which is an position where many moves have been played and maybe Fedorchuk's games as White are interesting for 8...a4 (see Chesspublishing updates). 

Here I propose the 8..h5!? pawn move